1 paper
Runduo Han, Weiming Xu, Zihan Zhang +2
The deep complex convolution recurrent network (DCCRN) achieves excellent speech enhancement performance by utilizing the audio spectrum's complex features. However, it has a large…